Table of Content

15 March 2022 Volume 35 Issue 3

  

Sparse Quantization Level Method for Scintillation Pulse Based on Feedback Regulation

Zhenzhou DENG,Qin HU,Wensheng LAI,Chunlei HAN,Ling TAO

Electronic Science and Technology. 2022, 35(3):  1-7.  doi:10.16180/j.cnki.issn1007-7820.2022.03.001

Abstract ( 183 )   HTML ( 32 )   PDF (2520KB) ( 62 )  

Figures and Tables | References | Related Articles | Metrics

The PET detector uses SQL sampling circuit to process scintillation pulse and fix the quantization level. The electric flipping caused by zero noise leads to abnormal heat of the voltage comparator for a long time, and the lower quantization level cannot be set, which limits the improvement of the detector's coincidence timing resolution. This study proposes a flicker pulse processing method based on feedback adjustment, which uses the switching characteristics of the transistor in the saturation state to perform feedback adjustment on the SQL quantitative samples. The delay element is used to delay and synchronize the quantized samples to generate a quantized level that can vary with the logic level of the sample, avoiding the influence of zero-point noise on the SQL sampling circuit. The experimental results show that compared with the original SQL sampling circuit, the optimized SQL sampling circuit reduces the amount of samples by 29% in the same time, and coincidence timing resolution is improved by 179.3ps, which reduces the useless power of PET detector and improves the coincidence timing resolution of PET system.

Speech Enhancement Method Based on Convolutional Recurrent Network and Non-Local Module

Hui LI,Hao JING,Kanghua YAN,Lianghao XU

Electronic Science and Technology. 2022, 35(3):  8-15.  doi:10.16180/j.cnki.issn1007-7820.2022.03.002

Abstract ( 220 )   HTML ( 1 )   PDF (2463KB) ( 30 )  

Figures and Tables | References | Related Articles | Metrics

The existing deep neural network speech enhancement methods ignore the importance of phase spectrum learning and cause the enhanced speech quality to be unsatisfactory. In view of this problem, a speech enhancement method based on convolutional recurrent network and non-local modules is proposed in the present study. By designing an encoder-decoder network, the time-domain representation of the speech signal is used as the input of the encoding end for deep feature extraction, so as to make full use of the amplitude information and phase information of the speech signal. Non-local modules are added to the convolutional layers of the encoder and decoder to extract key features of the speech sequence while suppressing useless features. A gated loop unit network is introduced to capture the timing correlation information between the speech sequences. The experimental results on the ST-CMDS Chinese speech dataset show that compared with the unprocessed noisy speech, the quality and intelligibility of the enhanced speech are improved by 61% and 7.93% on average.

Multi-Line Structured Light Centerline Extraction Method Based on System Space Structure Constraints

Xueyi MAO,Wenguo LI

Electronic Science and Technology. 2022, 35(3):  16-24.  doi:10.16180/j.cnki.issn1007-7820.2022.03.003

Abstract ( 281 )   HTML ( 3 )   PDF (3040KB) ( 63 )  

Figures and Tables | References | Related Articles | Metrics

The linear structured light measurement system is easy to be disturbed by ambient light, which affects the extraction of the center line of the fringe. Based on the analysis of the spatial structure constraints of the system, this study presents a method to extract the center line of multi-line structural light stripes. The extraction method is divided into two stages: calibration and measurement. In the calibration phase, the offset and offset coefficient of the center line are calculated. In the measurement stage, the three-dimensional world coordinates of the stripe center line are obtained with and without ambient light, respectively. The experimental accuracy is measured by plane and surface fitting and compared with Steger algorithm. The experimental results show that in the presence of ambient light, the error of skew plane fitting is less than 0.007 mm, and the percentage relative error of surface fitting is less than 1.2%. In the absence of ambient light, the relative error of cylinder height fitting is less than 0.7%. The above results indicate that the accuracy of the proposed algorithm is better than that of the Steger algorithm.

Early Warning and Monitoring Technology of Marine Internal Wave Based on Acoustic Vector Field Processing

Yu JIANG,Min ZHANG,Xingyu BAI,Shenghui HUA

Electronic Science and Technology. 2022, 35(3):  25-31.  doi:10.16180/j.cnki.issn1007-7820.2022.03.004

Abstract ( 219 )   HTML ( 3 )   PDF (1432KB) ( 41 )  

Figures and Tables | References | Related Articles | Metrics

In practical applications, the current ocean internal wave detection technology has large errors, and is significantly affected by the marine environment, and cannot be identified independently. In view of these problems, this study proposes a monitoring method for ocean internal wave early warning based on vector field processing. This method is based on the combined information processing of sound pressure and vibration velocity, and uses the three-dimensional information of the sound field picked up by the ultra-low frequency vector hydrophone. The time-space-frequency three-dimensional tracking and locking of non-cooperative targets can be carried out in the complex ocean background noise field according to the azimuth estimation algorithm. The arrival of internal wave causes the change of the three-dimensional sound velocity profile. The fluctuation of the sound field will lead to the change of the acoustic energy flow intensity of the target signal source. This method realizes the monitoring and prediction of ocean internal waves based on the abnormal jump of the vertical grazing angle caused by the channel distortion of the target signal in the internal wave space. The simulation results show that the vertical grazing angle fluctuates slightly in a normal environment, and the range of change is small. When the internal wave strikes, the grazing angle changes strongly, and the maximum deflection can be abruptly changed to a negative angle, which proves the effectiveness of the method.

Research on Adaptive Backstepping Control of Quadrotor UAV

Xinge SHEN,Hai JIN,Liang GUO

Electronic Science and Technology. 2022, 35(3):  32-37.  doi:10.16180/j.cnki.issn1007-7820.2022.03.005

Abstract ( 766 )   HTML ( 32 )   PDF (730KB) ( 118 )  

Figures and Tables | References | Related Articles | Metrics

An adaptive controller based on backstepping is designed to solve the problem of attitude and position control stability of UAVs under external interference. The dynamic modeling for the "X" UAV is conducted, and then converted into a strict feedback form with external interference. The system is divided into a position subsystem and an attitude subsystem. Combined with backstepping control, Lyapunov function is used to recursively deduce the adaptive law and the control law of each channel to make the system stable. Then the desired attitude angle can be inversely obtained through the channel control law, so that the system forms a closed loop. The simulation results of MATLAB-Simulink simulation model show that the controller designed in this study can effectively track the attitude and position of the UAV well in the presence of external interference.

CTA Segmentation Algorithm of Abdominal Artery Based on 3D Fully Convolutional Network

Lingyu JI,Yongbin GAO,Chenglu ZHAO,Xianhua TANG,Kaicheng XU,Jiacheng XU

Electronic Science and Technology. 2022, 35(3):  38-44.  doi:10.16180/j.cnki.issn1007-7820.2022.03.006

Abstract ( 277 )   HTML ( 2 )   PDF (1972KB) ( 44 )  

Figures and Tables | References | Related Articles | Metrics

Convolutional neural networks have become a research hotspot in the field of abdominal artery segmentation. The classic convolutional network has the problems of low segmentation accuracy and discontinuous segmentation of blood vessels. In view of these problems, this study proposes an abdominal arterial vessel segmentation algorithm based on an improved 3D full convolutional network. The side input of different scales is constructed on the encoding path of the network, and the convoluted image of side input is fused with the convoluted image of down sampling to extract more feature information. Meanwhile, a new multi-scale feature extraction module is embedded in the network. In this module, the channel attention and dense dilation convolution are introduced to capture the higher-level feature information. The experimental results on abdominal artery segmentation show that compared with other segmentation methods, the proposed method is more intuitive and quantitative, indicating that this method can improve the accuracy of blood vessel segmentation.

Arrhythmia Recognition Based on GAN-CNN

Peng CHEN,Zilong LIU

Electronic Science and Technology. 2022, 35(3):  45-50.  doi:10.16180/j.cnki.issn1007-7820.2022.03.007

Abstract ( 437 )   HTML ( 11 )   PDF (1573KB) ( 82 )  

Figures and Tables | References | Related Articles | Metrics

ECG analysis is an important basis for doctors to diagnose arrhythmia. The judgment of arrhythmia helps patients understand their physical conditions in time and find potential diseases. However, ECG analysis is not only time-consuming and labor-intensive, but also relies on clinical experience. Therefore, the efficiency of ECG analysis has always been limited by the number of doctors and work efficiency. The development of deep learning technology provides a foundation for the development of computer-aided diagnosis systems. In this study, a one-dimensional ECG signal is converted into a two-dimensional gray image, and a GAN-CNN network is used to solve the problem of ECG data imbalance, which can simultaneously realize the recognition of 7 types of arrhythmia and normal heartbeat. The experiment is verified by the MIT-BIH arrhythmia database. The average accuracy rate reaches 99.32%, and the sensitivity and specificity are 99.69% and 98.91%, respectively.

Prediction of PM_2.5 Based on External Influences and Time-Series Factors

Yanmei YANG,Zongmao CHENG

Electronic Science and Technology. 2022, 35(3):  51-57.  doi:10.16180/j.cnki.issn1007-7820.2022.03.008

Abstract ( 208 )   HTML ( 3 )   PDF (2079KB) ( 38 )  

Figures and Tables | References | Related Articles | Metrics

As the haze problem gradually worsens, the prediction of one of its main component PM_2.5 has become a widespread concern. The daily concentration of PM_2.5 is affected by many factors, and it has the characteristics of non-linear and time-varying, which is difficult to accurately predict.To solve this problem, a prediction method of PM_2.5 daily concentration based on external influences and time-series factors is proposed. With this method, the main external factors and time factors of PM_2.5 daily concentration are separated, and the BP neural network preliminary prediction model based on the main external factors and the combined residual correction model of EEMD-LSTM neural network based on time factor are established. The daily PM_2.5 concentration and other related factors data of Hangzhou from 2014 to 2019 are used for simulation experiments. The results show that compared with other models, the root mean square error of the prediction model proposed in the study is 2.74, and the prediction accuracy is higher.

Amplitude and Phase Error Estimation Algorithm for Sparse Planar Array

Wanying XIE,Hongjun JIN,Huaisong ZHAO

Electronic Science and Technology. 2022, 35(3):  58-64.  doi:10.16180/j.cnki.issn1007-7820.2022.03.009

Abstract ( 203 )   HTML ( 3 )   PDF (2084KB) ( 31 )  

Figures and Tables | References | Related Articles | Metrics

The sparse array has ambiguity characteristics, which makes the traditional ISM method unable to be directly applied when the amplitude and phase errors or position errors of the array elements exist. In view of this defect, the study utilizes the principle of DOA estimation algorithm based on medium defuzzification in linear arrays, extends this algorithm to two-dimensional sparse planar arrays, and deeply studies the influence of medium parameters on the suppression effect by introducing the concept of suppression degree. This study combines the algorithm with the classic ISM method to establish a new sparse planar array amplitude and phase error estimation algorithm. The simulation results show that this proposed method can not only make the blur angle suppression degree up to 17 dB under the condition of reasonable selection of media parameters, but also can estimate the amplitude and phase error under the premise of obtaining the accurate angle, which improves the robustness.

Design and Implementation of GPU Configuration Management System Based on WinForm

Zhao NIE,Jiawen HE,Chengcheng MA,Hui LIU

Electronic Science and Technology. 2022, 35(3):  65-70.  doi:10.16180/j.cnki.issn1007-7820.2022.03.010

Abstract ( 177 )   HTML ( 5 )   PDF (2140KB) ( 58 )  

Figures and Tables | References | Related Articles | Metrics

There are many GPU function configuration modules, the setting process steps are cumbersome, and the relationship between the configuration items is close, which increases the difficulty of GPU application development. In view of this problem, a GPU configuration management system based on WinForm is designed. The system is divided into software data generation and hardware environment configuration, which has the functions of completing startup/alarm screen generation, drive configuration, assembly tool, cursor generation, interface configuration, and font tool. According to the configuration information provided by the users, the algorithm automatically configures the module to initialize the content and sequence of steps. Users do not need to use the hard-coded form of function calls, and only use the mouse and keyboard to complete complex GPU configuration tasks. The function, design and implementation of the system are verified based on the FPGA verification platform. The results show that the system simplifies the configuration process and shortens the configuration time under the premise of satisfying functional correctness, design completeness and implementation robustness.

Design Analysis and Modeling Simulation of Brushless DC Motor

Xuanfeng SHANGGUAN,Tingyu YANG,Jinsong WEI,Yongjian LIU

Electronic Science and Technology. 2022, 35(3):  71-78.  doi:10.16180/j.cnki.issn1007-7820.2022.03.011

Abstract ( 835 )   HTML ( 27 )   PDF (1749KB) ( 160 )  

Figures and Tables | References | Related Articles | Metrics

Research on the structure design and driving mode of single-phase brushless DC motor is carried out in this study. The basic equation of the motor under ideal working conditions is derived, and the design scheme of the motor is determined according to the actual needs of the project and combined with the design principles of the brushless DC motor. A single phase brushless DC motor with rated power of 38 W and rated speed of 750 rpm is designed using the summarized design scheme. The effect of the gradient air gap on the starting performance and the torque of the gear groove is analyzed, and the optimal air gap length is determined. By comparing the advantages and disadvantages of unipolar winding and bipolar winding form, the winding form is determined, and the number of turns is determined by combining the traditional motor design formula. The rationality of the design scheme is verified by the finite element method. According to the dynamic mathematical model of the motor, the model of the motor system is established in the Simulink environment, and the curves of the motor speed and torque are obtained by simulation. The simulation results are consistent with the theoretical analysis, which verifies the rationality of the motor design scheme and the validity of the motor model.

Design of Reliability Evaluation System of Traction Motor Rolling Bearing Based on MATLAB App Designer

Meiyi QI,Aihua LIAO

Electronic Science and Technology. 2022, 35(3):  79-86.  doi:10.16180/j.cnki.issn1007-7820.2022.03.012

Abstract ( 179 )   HTML ( 2 )   PDF (2681KB) ( 73 )  

Figures and Tables | References | Related Articles | Metrics

Due to the uneven load and harsh operating environment during the operation of rolling bearings of traction motors, it is necessary to evaluate the reliability of rolling bearings of traction motors for metro vehicles. Based on MATLAB App Designer, the reliability evaluation system of traction motor rolling bearing is designed and developed. By extracting the time-domain and frequency-domain feature indexes of the vibration signal, the principal component analysis method is used for feature fusion. The fused characteristic index is used as the response covariate of the Weibull proportional hazard model. The moth-fighting optimization algorithm is used to optimize the Weibull proportional hazard model parameters, and the reliability function is substituted into the reliability function to calculate the reliability at any time. MATLAB App Designer is used to create module interfaces for data entry, data analysis and reliability evaluation. Experimental results show that the designed reliability evaluation system of traction motor rolling bearing has the characteristics of friendly interface, easy operation and good reliability evaluation effect, and can be widely used in different fields.